Repair of Functional Tricuspid Regurgitation:
Comparison Between Suture Annuloplasty
ADULT CARDIAC
and Rings Annuloplasty
Xinsheng Huang, PhD, Chengxiong Gu, MD, Xu Men, MD, Jianqun Zhang, MD,
Bin You, MD, Hongjia Zhang, PhD, Hua Wei, MD, and Jingxing Li, PhD
Cardiac Surgery Department, Beijing Anzhen Hospital, Capital Medical University, Beijing, China
Background. The purpose of this study was to review
our experience with modified De Vega tricuspid annu-
loplasty versus ring annuloplasty for treating functional
tricuspid regurgitation (TR).
Methods. In all, 448 consecutive patients undergoing
tricuspid annuloplasty with concomitant procedures be-
tween 2000 and 2012 were included. Modified De Vega
annuloplasty was performed in 216 patients (group 1) and
ring annuloplasty in 232 patients (group 2). Clinical and
echocardiographic follow-up results were used to assess
TR grade. Recurrent TR was defined as grade 2 or greater
at echocardiography.
Results. Preoperative clinical and echocardiographic
characteristics were comparable in the two groups. Early
mortality was similar (group 1, 0.9%, versus group 2, 1.3%;
p < 0.67). Follow-up was available for 97%; New York
Heart Association class and symptoms of right-side heart
failure were significantly improved. Survival rates at 1 and
lthough both rigid and flexible system provide
A acceptable results for tricuspid valve (TV) repair, the
use of a rigid ring significantly increases the risk of early
annular dehiscence [1, 2]. A possible explanation is that
increased shearing forces are being experienced at the
septal portion of the annulus, and that these shearing
forces would be even higher when a rigid ring is used [1,
2]. This line of argument finds support in the studies
addressing the dynamics of the native tricuspid annulus
[3, 4]. There are significant changes in the tricuspid
annular dimensions from systole to diastole, with a 19%
change of annular circumference and a 30% change in the
annular area [4]. In addition, Ton-Nu and coworkers [5]
demonstrated that tricuspid dilation occurs mainly in
the free wall of the tricuspid annulus at the middle part
of the anterior leaflet and extending to the septal part of
the septal leaflet.
On the basis of this concept, beginning in 2000, we
developed a new method of TV repair with a modified De
Vega’s suture annuloplasty. This method, however, has
since been considered to remodel annular structures by
Accepted for publication Oct 11, 2013.
Address correspondence to Dr Gu, Beijing Anzhen Hospital, Anding Rd 2,
Beijing 100029, China; e-mail: gucabg@126.com.
Ó 2014 by The Society of Thoracic Surgeons
Published by Elsevier Inc
5 years were comparable (97% and 84% for group 1, and
96% and 82% for group 2, respectively). Postoperative
echocardiography showed significant improvement in
tricuspid valve function, with reduction in TR grade from
3.4 to 0.6, and no differences between groups. However,
recurrence-free survival was better for group 2 than for
group 1 (78.8% versus 74.5%; p < 0.62). Risk factors for
recurrent TR included severity of preoperative TR, atrial
fibrillation, and pulmonary hypertension.
Conclusions. The modified De Vega tricuspid annulo-
plasty is acceptable for repair of functional TR and im-
provements in clinical and echocardiographic status on a
long-term basis, although the long-term recurrence-free
survival appeared to be lower than that for ring
annuloplasty.
(Ann Thorac Surg 2014;97:1286–93)
Ó 2014 by The Society of Thoracic Surgeons
maintaining a trileaflet valve that is more physiologic, to
maintain flexibility of right ventricular (RV) pumping
action, and to prevent redilation. In this study, we review
and compare the long-term clinic and echocardiographic
results of the modified De Vega and ring annuloplasty for
repair of functional tricuspid regurgitation (TR) to deter-
mine the efficacy and durability of tricuspid annuloplasty.
In addition, we also analyzed the risk factors for recur-
rence of TR.
Material and Methods
Patient Characteristics
Between January 2000 and April 2012, 448 consecutive
patients with functional TR underwent tricuspid annulo-
plasty with mitral valve replacement and mitral-aortic
valve replacement at our institution. Patients with
concomitant procedures such as mitral valve repair or
with endocarditis or significant organic disease of the
tricuspid valve leaflets or patients with congenital
anomalies were excluded. The TR was always secondary
to right ventricular dilation or dysfunction caused by
mitral valve rheumatic disease in 358 patients (79.9%),
mitral and aortic valves rheumatic disease in 87 (19.4%),
and mitral prosthesis dysfunction in 3 (0.7%). Indications
0003-4975/$36.00
http://dx.doi.org/10.1016/j.athoracsur.2013.10.037
Ann Thorac Surg
2014;97:1286–93
for TV surgery were symptomatic severe TR in 409
(91.3%) patients or else asymptomatic moderate TR or
marked tricuspid annular dilation (>4.0 cm) in the pres-
ence of other indications for cardiac surgery in 39 (8.7%).
Of these, the main TV disease was annulus dilation in all
patients, combined with restriction of the leaflets in 12
(2.7%) patients, and rolled up and thickened leaflets in 3
(0.7%) patients.
All patients were prospectively entered into our patient
data management system and then retrospectively
analyzed. Our Institutional Review Board approved this
study and waived the need for individual consent.
Surgical Procedure
All annuloplasties were performed during concomitant
mitral or aortic valve surgery with cardioplegic arrest.
Three surgeons performed operations during the study
periods (G.C.X., M.X., Z.J.Q.). The choice of tricuspid
annuloplasty methods was selected by the surgeons’
preference: in group 1, modified De Vega’s suture
annuloplasty was performed in 216 patients; in group 2, a
flexible ring (Cosgrove-Edwards annuloplasty system;
Edwards Lifesciences, Irvine, CA) was implanted in 147
patients and a rigid ring (Edwards Lifesciences) in 85
patients. Ring annuloplasty was performed by standard
operative techniques [6]. Modified De Vega’s suture
annuloplasty was performed by placing a doubled
pledget-supported suture of 3-0 polypropylene from the
level of the septal annulus in front of the coronary sinus
to the midportion of the annulus of the anterior leaflet
(Fig 1). The pledget was placed on the RV surface of
the septal annulus; the other pledget was placed at
the atrial surface of the anterior annulus. This double
multifilament suture is then tied down snugly over a
Hegar dilator calibrated to 2 mm to 3 mm larger than
the predicted appropriate pulmonary annulus size
derived from published nomograms [7], satisfactorily
reducing the orifice size without placing undue strain
on the annuloplasty repair. In all patients, intraoperative
transesophageal echocardiography was performed to
confirm elimination of TR.
HUANG ET AL 1287
SUTURE AND RINGS ANNULOPLASTY FOR FUNCTIONAL TR
Echocardiography Assessment of Tricuspid
Annuloplasty
Serial transthoracic echocardiographic reports were used
to assess TR grade. Recurrent TR was defined by grade 2
or greater at echocardiography. The severity of TR was
ADULT CARDIAC
characterized as trivial (grade 1, jet area/right atrial
area, less than 10%); mild (grade 2, jet area/right atrial
area, 10% to 20%); moderate (grade 3, jet area/right
atrial area, 20% to 33%); and severe (grade 4, jet area/right
atrial area, 33% or more) [8]. All echocardiographic
examinations were performed by the same two cardiologists.
We did not assess intraobserver or interobserver variability
in estimating severity of TR, but accepted the readings
used for clinical decision making.
Follow-Up
Follow-up data were obtained from hospital chart reviews
and telephone interviews with patients or family mem-
bers, When follow-up was not possible, information on
vital status (alive or dead) was obtained through the social
security database. Follow-up was closed on April 30, 2013.
We described postoperative events and results according
to the guidelines for reporting mortality and morbidity
after cardiac valve interventions, approved by The Society
of Thoracic Surgeons [9]. Every patient underwent clinical
assessment by a physician who was blinded to the
treatment assigned. Deaths and redo valve-related sur-
gery were recorded. Patients’ records included age, sex,
findings of physical examination (cyanosis, jaundice, neck
vein engorgement, ascites, hepatomegaly, and pitting
edema), and presence of atrial fibrillation on electrocar-
diogram. Follow-up was 96.7% complete, with a median
follow-up of 7.4 years (range, 6 months to 12 years).
Statistics Analysis
Preoperative variables were compared between groups
by the Fisher exact test (categorical variables) and the
Student t test (continuous variables). Kaplan-Meier
analysis was used to evaluate mortality and develop-
ment of 2þ (moderate) recurrent TR. Univariate analysis
was performed using the Breslow test and log rank test to
Fig 1. Schematic drawing of modification of
De Vega’s technique. (A) Operative tech-
nique. Modified De Vaga annuloplasty is
performed by placement of a 3–0 pledget-
supported polypropylene suture from the
posteroseptal commissures to the midportion
of anterior leaflet along the posterior annulus.
(B) Operative technique (continued). The
suture was tied down snugly over a Hegar
dilator calibrated to 2 to 3 mm larger than the
predicted appropriate pulmonary annulus
size. (A ¼ anterior leaflet; CS ¼ coronary
sinus; P ¼ posterior leaflet; S ¼ septal leaflet.)
 1288 HUANG ET AL Ann Thorac Surg
SUTURE AND RINGS ANNULOPLASTY FOR FUNCTIONAL TR 2014;97:1286–93

Table 1. Patient Preoperative Characteristics using SPSS software version 12.0 (SPSS, Chicago, IL).
Probability values of 0.05 or less were considered statis-
Group 1 Group 2 p
Variable (n ¼ 216) (n ¼ 232) Value tically significant.

Age, years 62.7  12.4 64.2  13.2 0.56

ADULT CARDIAC

Hypertension 76 (35.2) 83 (35.8) 0.09

Atrial fibrillation 198 (91.7) 208 (89.7) 0.54
Results
NYHA class 0.11 Perioperative Characteristics
IV 78 (36.1) 96 (41.3) Table 1 shows the demographic data of all patients. More
III 96 (44.4) 112 (48.3) than half of the patients had atrial fibrillation and 80% of
II 42 (19.4) 24 (10.3) patients were in New York Heart Association (NYHA)
Physical examination 0.12 functional class III or IV. There was no significant
Cyanosis 198 (91.7) 215 (92.7) difference between the two groups of patients in age,
Jaundice 89 (41.2) 114 (49.1) sex, NYHA class, preoperative pulmonary artery systolic
Neck vein engorgement 203 (94.0) 221 (95.2) pressure (PASP), preoperative RV dysfunction, and
Ascites 86 (40.0) 102 (44.0) concomitant surgical procedures. There were higher
Hepatomegaly 186 (86.1) 206 (88.8) incidences of symptoms of right-side heart failure, such
Pitting edema 146 (67.6) 176 (75.9) as neck vein distention, hepatomegaly, and pretibial
pitting edema, and lower left ventricular ejection fraction
Data are mean  SD or number (%). A p value less than 0.05 was (LVEF) in group 2. Table 2 also shows that there was a
considered significant.
higher percentage of patients with 3þ to 4þ TR in the
NYHA ¼ New York Heart Association. ring annuloplasty group than in the De Vega
annuloplasty group (93.0% versus 88.2%; p < 0.23), and
the mean TR was worse in the ring group (3.6 versus
determine whether any of the collected variables were 3.4; p < 0 .14).
predictors of late mortality. Multivariate analysis was Table 3 shows perioperative characteristics. The rate of
performed using Cox regression as well. Survival was mitral valve replacement (80% in group 1, 81% in group 2;
expressed using Kaplan-Meier curves. Midterm survival p ¼ 0 .75), and Maze procedure (60% versus 61%, p ¼ 0.37)
and freedom from morbid events (event-free survival were similarly performed in both treatment groups.
status) were compared using the Kaplan-Meier method; Operative time averaged 4 hours in all cases. With
independent predictors were determined by means of regard to mean cardiopulmonary bypass time and aortic
Cox regression analysis. Statistical data were analyzed cross-clamp time, the times in group 2 (200  82 and

Table 2. Perioperative and Follow-Up Echocardiographic Characteristics

Group 1 (n ¼ 216) Group 2 (n ¼ 232)

Variable Baseline At Discharge Follow-Up Baseline At Discharge Follow-Up

NYHA class 2.6  0.8 1.4  0.8 0.7  0.6 a

2.8  0.7 1.6  0.5 0.6  0.5a
Left atrial diameter, mm 60  10 48  9 45  4 67  8 50  6 47  7
LV end-diastolic diameter, mm 63  7 54  5 51  6a 65  9 55  8 52  7a
LV end-systolic diameter, mm 45  9 42  7 39  7a 45  11 42  9 41  8a
LV ejection fraction, % 43  12 52  14 56  13a 45  8 54  7 57  8a
<0.40 156 (72.2) 58 (26.9) 12 (5.6) 186 (80.1) 75 (32.3) 18 (7.6)a
RV long axis, mm 72  7 62  6 48  5a 73  9 67  8 47  11a
RV short axis, mm 36  4 32  5 28  6a 39  7 29  5 27  9a
TR grade
4þ 124 (57.4) 0 (0) 22 (10) 145 (62.5) 0 (0) 18 (8)
3þ 68 (31.4) 0 (0) 15 (7) 72 (31.0) 2 (0.8) 14 (6)
2þ 16 (7.4) 12 (5.6) 30 (14) 10 (4.3) 14 (6.0) 26 (11)
1þ 8 (3.7) 202 (93.5) 149 (69) 5 (2.2) 213 (93.0) 174 (75)
Mean TR grade 3.4  0.6 1.6  1.2a 0.8  0.3a 3.6  0.4 1.5  1.0a 0.7  0.4a
PASP, mm Hg 48  4 35  8 25  7a 49  4 36  6 23  5a
Transtricuspid gradient, mm Hg 28  13 23  9 18  6a 30  9 24  11 15  8a
a
Denotes p less than 0.05, baseline versus follow-up.
Data are mean  SD, or number (%).
EF ¼ ejection fraction; LV ¼ left ventricular; NYHA ¼ New York Heart Association; PASP ¼ pulmonary artery systolic pressure; RV ¼
right ventricular; TR ¼ tricuspid regurgitation.
Ann Thorac Surg HUANG ET AL 1289
2014;97:1286–93 SUTURE AND RINGS ANNULOPLASTY FOR FUNCTIONAL TR

Table 3. Perioperative Patient Characteristics Figure 2 shows the severity of TR before and after TV
repair. Most of the patients had moderate or severe TR
Variable Group 1 Group 2 p Value
preoperatively; however, 8.7% had mild or even no TR
Operative time, minutes 238.0  66.7 249.5  73.2 0.19 at the time of surgery. In these cases, the indication for
Cross-clamp time, minutes 121  72 148  79 0.58 TV repair was dilation of the native tricuspid annulus.

ADULT CARDIAC
CPB time, minutes 200  82 214  79 0.71 All annuloplasty types proved efficacious at reducing
Preoperative TR grade 2.8  0.6 3.0  0.6 0.06 TR. The TR was significantly reduced at discharge and
Postoperative TR grade 0.7  0.6 0.7  0.7 0.14 follow-up in all groups (p < 0.05), but it did increase at
Surgical procedure 0.44 follow-up controls in both groups (Table 2). In contrast, in
MVR 173 (80) 188 (84) 0.23 group 2 patients, it remained stable at further control
MVRþAVR 43 (20) 44 (18) 0.43 follow-up. Indeed, at late follow-up, TR in group 1 pa-
Maze procedure 130 (60) 142 (61) 0.37 tients was zero to mild in 69%, moderate in 14%, mod-
Mortality 2 (0.9) 3 (1.2) 0.67 erate to severe in 7%, and severe in 10%. In group 2
patients, TR was zero to mild in 75%, moderate in 11%,
Data are mean  SD or number (%). moderate to severe in 6%, and severe in 8%. There was no
AVR ¼ aortic valve replacement; CPB ¼ cardiopulmonary statistically significant difference in mean TR grade or
bypass; MVR ¼ mitral valve replacement; TR ¼ tricuspid prevalence of TR at last follow-up between the two
regurgitation.
groups (p < 0 .18). However, it got worse in group 2,
because the percentage of patients with TR equal to or
121  72 minutes, respectively) were longer than those in greater than grade 3 was higher in group 2 than in group
group 1 (148  79 minutes, respectively). Cardiopulmo- 1, but the distribution of TR grade was similar in the two
nary bypass and aortic crossclamp times were thus 20 groups.
minutes (p < 0.22) and 17 minutes (p < 0.17) shorter, To evaluate overall efficacy and durability of annulo-
respectively, in group 1. Controlling for number of plasty, we evaluated TR grades by the serial echocardio-
concomitant procedures, there was no statistically sig- graphic data. We determined TR grade at last follow-up
nificant difference in operative times (p < 0 .19). and assessed freedom from development of 2þ TR over
time using Kaplan-Meier analysis (Fig 3). There was no
significant difference between the two groups in the
Echocardiographic Assessment Kaplan-Meier analysis (p < 0.16 weighted by propensity
As expected, after surgical intervention, a reduction in left score; p < 0 .13 unweighted). Six-year freedom from
atrial and left ventricular dimensions was observed in both development of 2þ TR was 74.5%  6.6% and 78.8% 
groups, as well as in PASP, which significantly decreased 5.7% in group 1 and group 2, respectively.
from baseline in both groups (p < 0.05). Right ventricular
pressure also decreased significantly, but there was no Clinical Results
difference between the two groups (p ¼ 0.97; Table 2).
In each group, 3 patients (group 1, 1.4%, and group 2,
1.3%; p < 0.8) had a postoperative stroke; a total of
60 patients (group 1, 12%, and group 2, 14.6%; p < 0 .7)
had postoperative low cardiac output syndrome; and

Fig 2. Severity of tricuspid regurgitation (TR) before and after

tricuspid valve repair: (A) preoperative; (B) at discharge; (C) at Fig 3. Kaplan-Meier analysis of freedom from moderate tricuspid
follow-up. regurgitation (TR).
 1290 HUANG ET AL
SUTURE AND RINGS ANNULOPLASTY FOR FUNCTIONAL TR
8 patients (group 1, 1.3%, and group 2, 2.2%; p < 0 .9)
underwent reoperation because of postoperative bleeding.
Overall 30-day mortalities were 2 deaths (0.9%) in group 1
and 3 deaths (1.2%) in group 2 (p < 0 .07). Kaplan-Meier
survival rates at 1, 5, and 10 years were 97%, 84%, 76%,
ADULT CARDIAC
respectively, in group 1; and 96%, 82%, 72%, respectively,
in group 2 (Fig 4). There was no statistically significant
difference in mortality between the two groups;
however, patients of group 2 demonstrated a trend to
poorer survival (p < 0 .07). Interestingly, group 2 had a
lower preoperative ejection fraction and higher
preoperative TR than group 1.
A total of 10 patients underwent reoperation post-
operatively because of recurrent severe TR (group 1, 3
patients; group 2, 7 patients) during follow-up; the TV
reoperation rate in both groups was less than 2.2%. For
the 3 patients of group 1 who required reoperation, the
causes were central insufficiency in 2 and restrictive
leaflet motion in 1. For the 7 patients of group 2 requiring
reoperation, the causes were ring dehiscence in 4 pa-
tients, central insufficiency in 2, and restrictive leaflet
motion in 1 patient. Figure 5 reveals that 1-year and
5-year freedom from TV reoperation was 95%  2% and
81%  5%, respectively, in group 1, versus 93%  5%
and 75%  4 %, respectively, in group 2 (log rank test,
p ¼ 0.124).
The mean NYHA class improved from 2.6 to 0.6 in
group 1, and from 2.8 to 0.6 in group 2. Preoperatively,
44.4% of group 1 patients and 48.2% of group 2 patients
were in NYHA class III or IV. At follow-up, 7% of group 1
and 5% of group 2 patients remained in class III or IV.
Preoperative physical examination showed cyanosis in
413 patients (92.1%), jaundice in 203 (45.3%), neck vein
engorgement in 424 (94.6%), hepatomegaly in 392 (87.5%),
ascites in 186 (41.5%), and pitting edema in 322 (71.8%).
Postoperative physical examination found neck vein
engorgement in 15 patients (3.5%) and hepatomegaly in
18 (4.3%); no patient had ascites or pitting edema during
follow-up. There was a significant reduction of preoper-
ative symptoms of right-side heart failure.
Fig 4. Kaplan-Meier survival analysis.
Ann Thorac Surg
2014;97:1286–93
Fig 5. Postoperative freedom from tricuspid-related reoperation.
Risk Factors for Repair Failure
We evaluated the effect of preoperative and intraoperative
variables on the reoccurrence of TR after annuloplasty,
defined as grade 2 or higher. Follow-up echocardiograms
were available for 428 of the 443 patients (96.7%) during the
follow-up periods. A total of 5,190 postoperative echocar-
diographic reports were obtained. Univariate analysis
showed that preoperative hepatomegaly, neck vein
engorgement, LVEF of 0.40 or less, atrial fibrillation, PASP,
preoperative TR grade, left atrial dimension, and left ven-
tricular end-diastolic dimension were risk factors for re-
sidual TR after TV repair (p < 0.05). By multivariate binary
logistic regression analysis, higher preoperative pulmo-
nary artery pressure, higher preoperative TR grade, and
atrial fibrillation were found to be risk factors for devel-
opment of recurrent TR (Table 4). Significantly, tricuspid
annuloplasty type was not found to be a risk factor for
recurrent TR. Preoperative and follow-up NYHA class
were also not found to be significant factors.
Comment
In this study, we have evaluated TR over time after
modified De Vega’s suture annuloplasty and ring
tricuspid annuloplasty. Six-year freedom from develop-
ment of 2þ TR was 74.5%  6.6% and 78.8%  5.7% for
modified De Vega’s suture annuloplasty and ring annu-
loplasty, respectively; these results are similar to a pre-
vious study performed by Bernal and colleagues [10].
However, the long-term recurrence-free survival of su-
ture annuloplasty appeared to be lower than that of ring
annuloplasty.
Surgical Methods to Treat TR
There are two principal surgical methods to treat or
prevent TR: ring annuloplasty, as introduced by Car-
pentier and associates [11], and the suture annuloplasty
method, mostly performed as described by De Vega
and coworkers [12]. Although both rigid and flexible
Ann Thorac Surg
2014;97:1286–93
Table 4. Univariate and Multivariate Logistic Regression Analysis for Recurrent Tricuspid Regurgitation After Operation
Univariate Logistic Regression
Variable p Value OR
Hepatomegaly 0.013 2.667
Neck vein engorgement 0.017 2.385
PASP 0.023 3.353
LA dimension 0.012 2.561
Preoperative TR 0.000 5.433
Atrial fibrillation 0.034 3.638
Ejection fraction <0.40 0.002 4.563
NYHA class IV 0.002 5.412
LVEDD 0.033 3.267
CI ¼ confidence interval; LA ¼ left atrial; LVEDD ¼ left ventricular end-diastolic dimension;
odds ratio; PASP ¼ pulmonary artery systolic pressure; TR ¼ tricuspid regurgitation.
system provide acceptable results for TV repair, the use of
a rigid ring significantly increases the risk of early
annular dehiscence. Pfannm€ uller and colleagues [1], for
example, found that use of a Carpentier-Edwards ring
was associated with a significantly higher risk of dehis-
cence (Carpentier-Edwards, 8.7%; Cosgrove-Edwards,
0.9%; p < 0.001), almost exclusively at the septal leaflet
portion of the annulus. Multivariate analysis identified
annuloplasty type as independently predicting ring
dehiscence (odds ratio 10.7; 95% confidence interval: 3.2
to 36.5; p < 0.001). There is, therefore, a need to reassess
the early and medium-term postoperative outcomes in
patients undergoing tricuspid valve suture annuloplasty
with left-sided valve surgery in modern cardiac surgery.
Mechanism of the Modified De Vega Annuloplasty
The original De Vega annuloplasty and its variants purse
the posterior annular segment considerably more than the
anterior segment [13]. These suture annuloplasties, both
the continuous and single running types, have been
criticized for being unpredictable and unreliable, perhaps
because of the long suture line or the use of
polypropylene suture material, which may break and
slide through the tissue as the annulus dilates [13]. The
anatomic studies of Deloche and coworkers [14]
suggested that functional TR is related to the asymmetric
dilation of the tricuspid valve, which mainly affects the
posteroexternal half of the tricuspid circumference.
Furthermore, these changes involve both the anterior and
posterior annulus rather than the septum. Hence, the
ideal annuloplasty device should take into account the
geometric changes and restore or allow the normal three-
dimensional elliptical shape of the annulus to reduce
leaflet stress and possible tethering.
On the basis of this concept, as illustrated by Figure 1,
the modified De Vega annuloplasty plicated in
counterclockwise direction the junction of the annulus
and the right ventricle along the septal and posterior
leaflets and the midportion of the anulus of the anterior
leaflet, and then, tied down snugly over a Hegar dilator,
the suture does not cover the annulus from anteroseptal
commissure on the left to midseptal region (coronary
HUANG ET AL 1291
SUTURE AND RINGS ANNULOPLASTY FOR FUNCTIONAL TR
Multivariate Logistic Regression
95% CI p Value OR 95% CI
ADULT CARDIAC
1.380–9.745
1.146–4.962
1.709–6.579 0.001 7.446 2.321–18.432
1.842–4.731
1.442–6.413 0.001 8.743 2.232–23.652
1.082–12.228 0.0573 1.517 1.218–10.526
1.043–9.452
1.142–8.631
1.221–5.361
NYHA ¼ New York Heart Association; OR ¼
sinus) on the right, differently from traditional De Vega
repair. This procedure distributes the forces of
distension over two double sutures and acts specifically
on the anteroposterior segment, which is the most
dilated portion of the tricuspid annulus, thus creating
annular narrowing without excessive tension.
This modified De Vega technique was successfully
used by us for preventing recurrent TR in patients with
moderate to severe TR. At 5 years, 69.3% of patients in
this study had zero to mild TR. Thus, we believe that
tricuspid annuloplasty should be performed not only in
patients with moderate TR, but in all patients, even
those with severe TR, especially when an effective
annuloplasty technique can be performed rapidly and
reproducibly. In our experience, this technique has been
as effective as ring annuloplasty in the midterm post-
operative period.
A few technical points need to be mentioned regarding
the use of this technique. First, careful attention must be
paid to the pledget placed in the RV surface of the septal
annulus. Second, another pledget is placed in the atrial
surface of the anterior annulus. The two pledgets are
tightened between the tricuspid annulus along the base of
the posterior leaflet. Thus, this suture is less likely to
break, not come loose, not cut through the annulus, and
reduces the dilated anulus to an almost normal diameter
size. In addition, the modified De Vega technique main-
tains a trileaflet valve with a more physiologic, stabilized
annulus. Third, the suture was tied down snugly over a
Hegar dilator, to allow tight closure, which decreases the
likelihood of polypropylene suture material breaking and
sliding through the tissue as the annulus dilates, and
hence, late tricuspid stenosis.
Risk Factors for Recurrent TR
Well-known risk factors for residual TR were also found
in this study [15]. Risk factor analysis using logistic
regression revealed that pulmonary hypertension,
severity of preoperative TR, and atrial fibrillation were
found to be risk factors for the development of
recurrent TR. Significantly, tricuspid annuloplasty type
was not found to be a risk factor for recurrent TR.
 1292 HUANG ET AL
SUTURE AND RINGS ANNULOPLASTY FOR FUNCTIONAL TR
Clinical Assessment
Our present study has shown that at 1 year and 5 years,
Kaplan-Meier survival rates, freedom from reoperation,
and improvement in NYHA functional class showed no
significant differences between the two groups. Although
ADULT CARDIAC
echocardiographic TR improved significantly in the
modified De Vega group compared with the ring group,
all differences between the two groups were subclinical.
Currently, RV function is considered to be one of the most
important factors in midterm and long-term survival of
patients with valvular heart disease [16]. The modified
De Vega’s suture annuloplasty, however, has since
been considered as remodeling annular structures by
maintaining a more physiologic trileaflet valve, to
maintain flexibility of RV pumping action. Our present
study has shown that significant reverse RV remodeling
and an improvement of the preoperative symptoms of
right-side heart failure are results similar to those of the
Benedetto study [17]. The notion of preserving RV
function and tricuspid annular contraction is attractive
and needs to be further studied.
Study Limitations
Our study has several limitations. First, despite being a
study conducted in a single center, the choice of annu-
loplasty method was selected by an individual surgeon,
and with some of the surgical staff members being the
same over the years, there have been changes not only in
the population of patients but also in the diagnostic
methods, and especially in relation to techniques of
valvular heart surgery and myocardial protection. Sec-
ond, the patients were not randomized, creating an
inherent selection bias with regard to the annuloplasty
performed. Although the two groups were very similar,
the ring annuloplasty group had slightly worse preoper-
ative TR and LVEF. Finally, there were no definite criteria
regarding reduction in tricuspid annular size. Further
comparisons of TV repair methods should be carried out
in a prospective randomized manner.
In conclusion, the modified De Vega tricuspid annu-
loplasty is a simple, inexpensive option for addressing
functional TR and effective improvement in clinical and
echocardiographic status on a long-term basis.
References
1. Pfannm€uller B, Doenst T, Eberhardt K, Seeburger J,
Borger MA, Mohr FW. Increased risk of dehiscence after
INVITED COMMENTARY
The article by Huang and colleagues [1], comparing
tricuspid valve (TV) repair techniques, clearly demonstrates
the need for and outcomes of correcting tricuspid
regurgitation (TR) of all causes. They have collected
these data from over 400 patients over a 12-year period,
with a very good follow-up.
Ó 2014 by The Society of Thoracic Surgeons
Published by Elsevier Inc
Ann Thorac Surg
2014;97:1286–93
tricuspid valve repair with rigid annuloplasty rings. J Thorac
Cardiovasc Surg 2012;143:1050–5.
2. Navia JL, Nowicki ER, Blackstone EH, et al. Surgical man-
agement of secondary tricuspid valve regurgitation: annulus,
commissure, or leaflet procedure? J Thorac Cardiovasc Surg
2010;139:1473–82.
3. Fukuda S, Saracino G, Matsumura Y, et al. Three-dimen-
sional geometry of the tricuspid annulus in healthy subjects
and in patients with functional tricuspid regurgitation: a real-
time, 3-dimensional echocardiographic study. Circulation
2006;114(Suppl):I492–8.
4. Tei C, Pilgrim JP, Shah PM, Ormiston JA, Wong M. The
tricuspid valve annulus: study of size and motion in normal
subjects and in patients with tricuspid regurgitation. Circu-
lation 1982;66:665–71.
5. Ton-Nu TT, Levine RA, Handschumacher MD, et al. Geo-
metric determinants of functional tricuspid regurgitation:
insights from 3-dimensional echocardiography. Circulation
200611;114:143–149.
6. Cosgrove DM, Arcidi JM, Rodriguez L, Stewart WJ,
Powell K, Thomas JD. Initial experience with the Cosgrove-
Edwards annuloplasty system. Ann Thorac Surg 1995;60:
499–504.
7. Pacifico AD, Kirklin JW, Blackstone EH. Surgical manage-
ment of pulmonary stenosis in tetralogy of Fallot. J Thorac
Cardiovasc Surg 1977;74:382–95.
8. Chopra HK, Nanda NC, Fan P, et al. Can two-dimensional
echocardiography and Doppler color flow mapping identify
the need for tricuspid valve repair? J Am Coll Cardiol
1989;14:1266–74.
9. Akins CW, Miller DC, Turina MI, et al. Guidelines for
reporting mortality and morbidity after cardiac valve in-
terventions. Ann Thorac Surg 2008;85:1490–5.
10. Bernal JM, Gutierrez-Morlote J, Llorca J, San Jose JM,
Morales D, Revuelta JM. Tricuspid valve repair: an old dis-
ease, a modern experience. Ann Thorac Surg 2004;78:
2069–75.
11. Carpentier A, Deloche A, Dauptain J, et al. A new recon-
structive operation for correction of mitral and tricuspid
insufficiency. J Thorac Cardiovasc Surg 1971;61:1–13.
12. Kay JH, Maselli-Campagna G, Tsuji KK. Surgical treatment
of tricuspid insufficiency. Ann Surg 1965;162:53–8.
13. Sarraj A, Nuche JM, Domínguez L, et al. Adjustable
segmental tricuspid annuloplasty: technical advantages and
midterm results. Ann Thorac Surg 2009;87:1148–53.
14. Deloche A, Guerinon J, Fabiani JN, et al. [Anatomical study
of rheumatic tricuspid valve diseases: application to the
study of various valvuloplasties]. Ann Chir Thorac Car-
diovasc 1973;12:343–9.
15. Fukuda S, Gillinov AM, McCarthy PM, et al. Determinants
of recurrent or residual functional tricuspid regurgitation
after tricuspid annuloplasty. Circulation 2006;114(Suppl):
I582–7.
16. Nagel E, Stuber M, Hess OM. Importance of the right
ventricle in valvular heart disease. Eur Heart J 1996;17:
829–36.
17. Benedetto U, Melina G, Angeloni E, et al. Prophylactic
tricuspid annuloplasty in patients with dilated tricuspid
annulus undergoing mitral valve surgery. J Thorac Car-
diovasc Surg 2012;143:632–8.
They have compared two techniques: a modified De
Vega technique and a complete ring annuloplasty, which
was pioneered by Dr Carpentier of France. The operative
mortality and the long-term correction of TR in their
hands have produced enviable results. They are to be
complimented especially for the very low mortality for a
0003-4975/$36.00
http://dx.doi.org/10.1016/j.athoracsur.2013.10.080